ff_analysis_5

This model is a fine-tuned version of zdreiosis/ff_analysis_5 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0824
• F1: 0.9306
• Roc Auc: 0.9483
• Accuracy: 0.8137

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Roc Auc	Accuracy
No log	0.27	50	0.0846	0.9305	0.9476	0.8075
No log	0.55	100	0.1000	0.9070	0.9320	0.7484
No log	0.82	150	0.0945	0.9126	0.9349	0.7640
No log	1.1	200	0.0973	0.9119	0.9353	0.7764
No log	1.37	250	0.0880	0.9336	0.9504	0.8261
No log	1.65	300	0.0857	0.9246	0.9434	0.8043
No log	1.92	350	0.0844	0.9324	0.9488	0.8199
No log	2.2	400	0.0881	0.9232	0.9450	0.7888
No log	2.47	450	0.0875	0.9277	0.9462	0.8012
0.1226	2.75	500	0.0824	0.9306	0.9483	0.8137

Framework versions

• Transformers 4.15.0
• Pytorch 1.11.0+cu113
• Datasets 2.3.2
• Tokenizers 0.10.3